3D Multimodal Fusion Network with Disease-induced Joint Learning for Early Alzheimer's Disease Diagnosis.

Multimodal neuroimaging provides complementary information critical for accurate early diagnosis of Alzheimer's disease (AD). However, the inherent variability between multimodal neuroimages hinders the effective fusion of multimodal features. Moreover, achieving reliable and interpretable diagnoses in the field of multimodal fusion remains challenging. To address them, we propose a novel multimodal diagnosis network based on multi-fusion and disease-induced learning (MDL-Net) to enhance early AD diagnosis by efficiently fusing multimodal data. Specifically, MDL-Net proposes a multi-fusion joint learning (MJL) module, which effectively fuses multimodal features and enhances the feature representation from global, local, and latent learning perspectives. MJL consists of three modules, global-aware learning (GAL), local-aware learning (LAL), and outer latent-space learning (LSL) modules. GAL via a self-adaptive Transformer (SAT) learns the global relationships among the modalities. LAL constructs local-aware convolution to learn the local associations. LSL module introduces latent information through outer product operation to further enhance feature representation. MDL-Net integrates the disease-induced region-aware learning (DRL) module via gradient weight to enhance interpretability, which iteratively learns weight matrices to identify AD-related brain regions. We conduct the extensive experiments on public datasets and the results confirm the superiority of our proposed method. Our code will be available at: https://github.com/qzf0320/MDL-Net.

The Efficacy of Adalimumab in Children with Chronic Non-infectious Posterior Uveitis and Panuveitis: A Retrospective Cohort Study.

INTRODUCTION: This study aimed to assess the efficacy and safety of adalimumab in pediatric patients with chronic non-infectious posterior uveitis and panuveitis (not associated with juvenile idiopathic arthritis). METHODS: The medical records of children (< 18 years old) with chronic non-infectious posterior uveitis and panuveitis were collected and analyzed in this retrospective cohort study. Children were allocated to a conventional adalimumab-free treatment (CT) or adalimumab (ADA) group based on whether they additionally received adalimumab. RESULTS: In total, 69 children (138 eyes) were included, with 21 (42 eyes) and 48 (96 eyes) in the CT and ADA groups, respectively. During the average follow-up period of 24 months, the improvement in all ocular parameters (best-corrected visual acuity, intraocular inflammation, fluorescein angiography score) was better in the ADA group than in the CT group, except for changes in central macular thickness, which did not significantly differ between the groups. The mean time of first alleviation, which was after 1.03 ± 0.12 months of therapy, was earlier in the ADA group than in the CT group (2.30 ± 0.46 months). In the ADA group, 90.6% of children had remission within 3 months, and 47.9% had no relapse during follow-up. Cough and cold were the most common adverse events in the ADA group; however, the number of adverse events was similar between both the groups. CONCLUSIONS: Adalimumab was effective in the treatment of chronic noninfectious posterior uveitis and panuveitis in pediatric patients, and disease inactivity was accomplished in the majority of the patients, thereby improving visual outcomes and maintaining disease stability. Adverse events were limited and tolerable.

CircEYA3 aggravates intervertebral disc degeneration through the miR-196a-5p/EBF1 axis and NF-κB signaling.

Intervertebral disc degeneration (IDD) is a well-established cause of disability, and extensive evidence has identified the important role played by regulatory noncoding RNAs, specifically circular RNAs (circRNAs) and microRNAs (miRNAs), in the progression of IDD. To elucidate the molecular mechanism underlying IDD, we established a circRNA/miRNA/mRNA network in IDD through standardized analyses of all expression matrices. Our studies confirmed the differential expression of the transcription factors early B-cell factor 1 (EBF1), circEYA3, and miR-196a-5p in the nucleus pulposus (NP) tissues of controls and IDD patients. Cell proliferation, apoptosis, and extracellular mechanisms of degradation in NP cells (NPC) are mediated by circEYA3. MiR-196a-5p is a direct target of circEYA3 and EBF1. Functional analysis showed that miR-196a-5p reversed the effects of circEYA3 and EBF1 on ECM degradation, apoptosis, and proliferation in NPCs. EBF1 regulates the nuclear factor kappa beta (NF-кB) signalling pathway by activating the IKKß promoter region. This study demonstrates that circEYA3 plays an important role in exacerbating the progression of IDD by modulating the NF-κB signalling pathway through regulation of the miR196a-5p/EBF1 axis. Consequently, a novel molecular mechanism underlying IDD development was elucidated, thereby identifying a potential therapeutic target for future exploration.

Morphology-based deep learning enables accurate detection of senescence in mesenchymal stem cell cultures.

BACKGROUND: Cell senescence is a sign of aging and plays a significant role in the pathogenesis of age-related disorders. For cell therapy, senescence may compromise the quality and efficacy of cells, posing potential safety risks. Mesenchymal stem cells (MSCs) are currently undergoing extensive research for cell therapy, thus necessitating the development of effective methods to evaluate senescence. Senescent MSCs exhibit distinctive morphology that can be used for detection. However, morphological assessment during MSC production is often subjective and uncertain. New tools are required for the reliable evaluation of senescent single cells on a large scale in live imaging of MSCs. RESULTS: We have developed a successful morphology-based Cascade region-based convolution neural network (Cascade R-CNN) system for detecting senescent MSCs, which can automatically locate single cells of different sizes and shapes in multicellular images and assess their senescence state. Additionally, we tested the applicability of the Cascade R-CNN system for MSC senescence and examined the correlation between morphological changes with other senescence indicators. CONCLUSIONS: This deep learning has been applied for the first time to detect senescent MSCs, showing promising performance in both chronic and acute MSC senescence. The system can be a labor-saving and cost-effective option for screening MSC culture conditions and anti-aging drugs, as well as providing a powerful tool for non-invasive and real-time morphological image analysis integrated into cell production.

An aging-related immune landscape in the hematopoietic immune system.

BACKGROUND: Aging is a holistic change that has a major impact on the immune system, and immunosenescence contributes to the overall progression of aging. The bone marrow is the most important hematopoietic immune organ, while the spleen, as the most important extramedullary hematopoietic immune organ, maintains homeostasis of the human hematopoietic immune system (HIS) in cooperation with the bone marrow. However, the overall changes in the HIS during aging have not been described. Here, we describe a hematopoietic immune map of the spleen and bone marrow of young and old mice using single-cell sequencing and flow cytometry techniques. RESULTS: We observed extensive, complex changes in the HIS during aging. Compared with young mice, the immune cells of aged mice showed a marked tendency toward myeloid differentiation, with the neutrophil population accounting for a significant proportion of this response. In this change, hypoxia-inducible factor 1-alpha (Hif1α) was significantly overexpressed, and this enhanced the immune efficacy and inflammatory response of neutrophils. Our research revealed that during the aging process, hematopoietic stem cells undergo significant changes in function and composition, and their polymorphism and differentiation abilities are downregulated. Moreover, we found that the highly responsive CD62L + HSCs were obviously downregulated in aging, suggesting that they may play an important role in the aging process. CONCLUSIONS: Overall, aging extensively alters the cellular composition and function of the HIS. These findings could potentially give high-dimensional insights and enable more accurate functional and developmental analyses as well as immune monitoring in HIS aging.

Boundary uncertainty aware network for automated polyp segmentation.

Recently, leveraging deep neural networks for automated colorectal polyp segmentation has emerged as a hot topic due to the favored advantages in evading the limitations of visual inspection, e.g., overwork and subjectivity. However, most existing methods do not pay enough attention to the uncertain areas of colonoscopy images and often provide unsatisfactory segmentation performance. In this paper, we propose a novel boundary uncertainty aware network (BUNet) for precise and robust colorectal polyp segmentation. Specifically, considering that polyps vary greatly in size and shape, we first adopt a pyramid vision transformer encoder to learn multi-scale feature representations. Then, a simple yet effective boundary exploration module (BEM) is proposed to explore boundary cues from the low-level features. To make the network focus on the ambiguous area where the prediction score is biased to neither the foreground nor the background, we further introduce a boundary uncertainty aware module (BUM) that explores error-prone regions from the high-level features with the assistance of boundary cues provided by the BEM. Through the top-down hybrid deep supervision, our BUNet implements coarse-to-fine polyp segmentation and finally localizes polyp regions precisely. Extensive experiments on five public datasets show that BUNet is superior to thirteen competing methods in terms of both effectiveness and generalization ability.

Cytoplasmic Escape of Mitochondrial DNA Mediated by Mfn2 Downregulation Promotes Microglial Activation via cGas-Sting Axis in Spinal Cord Injury.

Neuroinflammation is associated with poor outcomes in patients with spinal cord injury (SCI). Recent studies have demonstrated that stimulator of interferon genes (Sting) plays a key role in inflammatory diseases. However, the role of Sting in SCI remains unclear. In the present study, it is found that increased Sting expression is mainly derived from activated microglia after SCI. Interestingly, knockout of Sting in microglia can improve the recovery of neurological function after SCI. Microglial Sting knockout restrains the polarization of microglia toward the M1 phenotype and alleviates neuronal death. Furthermore, it is found that the downregulation of mitofusin 2 (Mfn2) expression in microglial cells leads to an imbalance in mitochondrial fusion and division, inducing the release of mitochondrial DNA (mtDNA), which mediates the activation of the cGas-Sting signaling pathway and aggravates inflammatory response damage after SCI. A biomimetic microglial nanoparticle strategy to deliver MASM7 (named MSNs-MASM7@MI) is established. In vitro, MSNs-MASM7@MI showed no biological toxicity and effectively delivered MASM7. In vivo, MSNs-MASM7@MI improves nerve function after SCI. The study provides evidence that cGas-Sting signaling senses Mfn2-dependent mtDNA release and that its activation may play a key role in SCI. These findings provide new perspectives and potential therapeutic targets for SCI treatment.

Development of a regenerable dual-trigger tripedal DNA walker electrochemical biosensor for sensitive detection of microRNA-155.

Since microRNAs (miRNAs) are valuable biomarkers for disease diagnosis and prognosis, the pursuit of enhanced detection sensitivity through signal amplification strategies has emerged as a prominent focus in low-abundance miRNA detection research. DNA walkers, as dynamic DNA nanodevice, have gained significant attention for their applications as signal amplification strategies. To overcome the limitations of unipedal DNA walkers with a restricted signal amplification efficiency, there is a great need for multi-pedal DNA walkers that offer improved walking and signal amplification capabilities. Here, we employed a combination of catalytic hairpin assembly (CHA) and APE1 enzymatic cleavage reactions to construct a tripedal DNA walker, driving its movement to establish a cascade signal amplification system for the electrochemical detection of miRNA-155. The biosensor utilizes tumor cell-endogenous microRNA-155 and APE1 as dual-trigger for DNA walker formation and walking movement, leading to highly efficient and controllable signal amplification. The biosensor exhibited high sensitivity, with a low detection limit of 10 pM for microRNA-155, and successfully differentiated and selectively detected microRNA-155 from other interfering RNAs. Successful detection in 20 % serum samples indicates its potential clinical application. In addition, we harnessed strand displacement reactions to create a gentle yet efficient electrode regeneration strategy, to addresses the time-consuming challenges during electrode modification processes. We have successfully demonstrated the stability of current signals even after multiple cycles of electrode regeneration. This study showcased the high-efficiency amplification potential of multi-pedal DNA walkers and the effectiveness and versatility of strand displacement in biosensing applications. It opens a promising path for developing regenerable electrochemical biosensors. This regenerable strategy for electrochemical biosensors is both label-free and cost-effective, and holds promise for detecting various disease-related RNA targets beyond its current application.

USBDAN: Unsupervised Scale-aware and Boundary-aware Domain Adaptive Network for Gastric Tumor Segmentation.

Accurate segmentation of gastric tumors from computed tomography (CT) images provides useful image information for guiding the diagnosis and treatment of gastric cancer. Researchers typically collect datasets from multiple medical centers to increase sample size and representation, but this raises the issue of data heterogeneity. To this end, we propose a new cross-center 3D tumor segmentation method named unsupervised scale-aware and boundary-aware domain adaptive network (USBDAN), which includes a new 3D neural network that efficiently bridges an Anisotropic neural network and a Transformer (AsTr) for extracting multi-scale features from the CT images with anisotropic resolution, and a scale-aware and boundary-aware domain alignment (SaBaDA) module for adaptively aligning multi-scale features between two domains and enhancing tumor boundary drawing based on location-related information drawn from each sample across all domains. We evaluate the proposed method on an in-house CT image dataset collected from four medical centers. Our results demonstrate that the proposed method outperforms several state-of-the-art methods.

Integration of (S)-2,3-oxidosqualene enables E. coli to become Iron Man E. coli with improved overall tolerance.

BACKGROUND: While representing a model bacterium and one of the most used chassis in biomanufacturing, performance of Escherichia coli is often limited by severe stresses. A super-robust E. coli chassis that could efficiently tolerant multiple severe stresses is thus highly desirable. Sterols represent a featured composition that distinguishes eukaryotes from bacteria and all archaea, and play a critical role in maintaining the membrane integrity of eukaryotes. All sterols found in nature are directly synthesized from (S)-2,3-oxidosqualene. However, in E. coli, (S)-2,3-oxidosqualene is not present. RESULTS: In this study, we sought to introduce (S)-2,3-oxidosqualene into E. coli. By mining and recruiting heterologous enzymes and activation of endogenous pathway, the ability of E. coli to synthesize (S)-2,3-oxidosqualene was demonstrated. Further analysis revealed that this non-native chemical confers E. coli with a robust and stable cell membrane, consistent with a figurative analogy of wearing an "Iron Man's armor"-like suit. The obtained Iron Man E. coli (IME) exhibited improved tolerance to multiple severe stresses, including high temperature, low pH, high salt, high sugar and reactive oxygen species (ROS). In particular, the IME strain shifted its optimal growth temperature from 37 °C to 42-45 °C, which represents the most heat-resistant E. coli to the best of our knowledge. Intriguingly, this non-native chemical also improved E. coli tolerance to a variety of toxic feedstocks, inhibitory products, as well as elevated synthetic capacities of inhibitory chemicals (e.g., 3-hydroxypropionate and fatty acids) due to improved products tolerance. More importantly, the IME strain was effectively inhibited by the most commonly used antibiotics and showed no undesirable drug resistance. CONCLUSIONS: Introduction of the non-native (S)-2,3-oxidosqualene membrane lipid enabled E. coli to improve tolerance to various stresses. This study demonstrated the effectiveness of introducing eukaryotes-featured compound into bacteria for enhancing overall tolerance and chemical production.

Dense Swin Transformer for Classification of Thyroid Nodules.

As an early sign of thyroid cancer, thyroid nodules are the most common nodular lesions. As a non-invasive imaging method, ultrasound is widely used in the diagnosis of benign and malignant thyroid nodules. As there is no obvious difference in appearance between the two types of thyroid nodules, and the contrast with the surrounding muscle tissue is too low, it is difficult to distinguish the benign and malignant nodules. Therefore, a dense nodal Swin-Transformer(DST) method for the diagnosis of thyroid nodules is proposed in this paper. Image segmentation is carried out through patch, and feature maps of different sizes are constructed in four stages, which consider different information of each layer of features. In each stage block, a dense connection mechanism is used to make full use of multi-layer features and effectively improve the diagnostic performance. The experimental results of multi-center ultrasound data collected from 17 hospitals show that the accuracy of the proposed method is 87.27%, the sensitivity is 88.63%, and the specific effect is 85.16%, which verifies that the proposed algorithm has the potential to assist clinical practice.

Fused Brain Functional Connectivity Network and Edge-attention Graph Convolution Network for Fibromyalgia Syndrome Diagnosis.

Fibromyalgia syndrome (FMS) is a type of rheumatology that seriously affects the normal life of patients. Due to the complex clinical manifestations of FMS, it is challenging to detect FMS. Therefore, an automatic FMS diagnosis model is urgently needed to assist physicians. Brain functional connectivity networks (BFCNs) constructed by resting-state functional magnetic resonance imaging (rs-fMRI) to describe brain functions have been widely used to identify individuals with relevant diseases from normal control (NC). Therefore, we propose a novel model based on BFCN and graph convolutional network (GCN) for automatic FMS diagnosis. Firstly, a novel fused BFCN method is proposed by fusing Pearson's correlation (PC) and low-rank (LR) BFCN, which retains information and reduces data redundancy to construct BFCN. Then we combine the feature of BFCN with non-image information of subjects to obtain nodes and adjacency matrices, which builds a graph with edge attention. Finally, the graph is sent to the GCN layer for FMS diagnosis. Our model is evaluated on the in-house FMS dataset to achieve 82.48% accuracy. The experimental results show that our method outperforms the state-of-the-art competing methods.

A Two-stage Diagnostic Framework for Post-ablation Treatment Response Assessment in Patients with Hepatocellular Carcinoma.

Contrast-enhanced ultrasound (CEUS) video plays an important role in post-ablation treatment response assessment in patients with hepatocellular carcinoma (HCC). However, the assessment of treatment response using CEUS video is challenging due to issues such as high inter-frame data repeatability, small ablation area and poor imaging quality of CEUS video. To address these issues, we propose a two-stage diagnostic framework for post-ablation treatment response assessment in patients with HCC using CEUS video. The first stage is a location stage, which is used to locate the ablation area. At this stage, we propose a Yolov5-SFT to improve the location results of the ablation area and a similarity comparison module (SCM) to reduce data repeatability. The second stage is an assessment stage, which is used for the evaluation of postoperative efficacy. At this stage, we design an EfficientNet-SK to improve assessment accuracy. The Experimental results on the self-collected data show that the proposed framework outperforms other selected algorithms, and can effectively assist doctors in the assessment of post-ablation treatment response.

Graph Convolution and Self-attention Enhanced CNN with Domain Adaptation for Multi-site COVID-19 Diagnosis.

In 2019, coronavirus disease (COVID-19) is an acute disease that can rapidly develop into a very serious state. Therefore, it is of great significance to realize automatic COVID-19 diagnosis. However, due to the small difference in the characteristics of computed tomography (CT) between community acquire pneumonia (CP) and COVID-19, the existing model is unsuitable for the three-class classifications of healthy control, CP and COVID-19. The current model rarely optimizes the data from multiple centers. Therefore, we propose a diagnosis model for COVID-19 patients based on graph enhanced 3D convolution neural network (CNN) and cross-center domain feature adaptation. Specifically, we first design a 3D CNN with graph convolution module to enhance the global feature extraction capability of the CNN. Meanwhile, we use the domain adaptive feature alignment method to optimize the feature distance between different centers, which can effectively realize multi-center COVID-19 diagnosis. Our experimental results achieve quite promising COVID-19 diagnosis results, which show that the accuracy in the mixed dataset is 98.05%, and the accuracy in cross-center tasks are 85.29% and 87.53%.

Adaptive Cross-Feature Fusion Network With Inconsistency Guidance for Multi-Modal Brain Tumor Segmentation.

In the context of contemporary artificial intelligence, increasing deep learning (DL) based segmentation methods have been recently proposed for brain tumor segmentation (BraTS) via analysis of multi-modal MRI. However, known DL-based works usually directly fuse the information of different modalities at multiple stages without considering the gap between modalities, leaving much room for performance improvement. In this paper, we introduce a novel deep neural network, termed ACFNet, for accurately segmenting brain tumor in multi-modal MRI. Specifically, ACFNet has a parallel structure with three encoder-decoder streams. The upper and lower streams generate coarse predictions from individual modality, while the middle stream integrates the complementary knowledge of different modalities and bridges the gap between them to yield fine prediction. To effectively integrate the complementary information, we propose an adaptive cross-feature fusion (ACF) module at the encoder that first explores the correlation information between the feature representations from upper and lower streams and then refines the fused correlation information. To bridge the gap between the information from multi-modal data, we propose a prediction inconsistency guidance (PIG) module at the decoder that helps the network focus more on error-prone regions through a guidance strategy when incorporating the features from the encoder. The guidance is obtained by calculating the prediction inconsistency between upper and lower streams and highlights the gap between multi-modal data. Extensive experiments on the BraTS 2020 dataset show that ACFNet is competent for the BraTS task with promising results and outperforms six mainstream competing methods.

Semi-Supervised Representation Learning for Segmentation on Medical Volumes and Sequences.

Benefiting from the massive labeled samples, deep learning-based segmentation methods have achieved great success for two dimensional natural images. However, it is still a challenging task to segment high dimensional medical volumes and sequences, due to the considerable efforts for clinical expertise to make large scale annotations. Self/semi-supervised learning methods have been shown to improve the performance by exploiting unlabeled data. However, they are still lack of mining local semantic discrimination and exploitation of volume/sequence structures. In this work, we propose a semi-supervised representation learning method with two novel modules to enhance the features in the encoder and decoder, respectively. For the encoder, based on the continuity between slices/frames and the common spatial layout of organs across subjects, we propose an asymmetric network with an attention-guided predictor to enable prediction between feature maps of different slices of unlabeled data. For the decoder, based on the semantic consistency between labeled data and unlabeled data, we introduce a novel semantic contrastive learning to regularize the feature maps in the decoder. The two parts are trained jointly with both labeled and unlabeled volumes/sequences in a semi-supervised manner. When evaluated on three benchmark datasets of medical volumes and sequences, our model outperforms existing methods with a large margin of 7.3% DSC on ACDC, 6.5% on Prostate, and 3.2% on CAMUS when only a few labeled data is available. Further, results on the M&M dataset show that the proposed method yields improvement without using any domain adaption techniques for data from unknown domain. Intensive evaluations reveal the effectiveness of representation mining, and superiority on performance of our method. The code is available at https://github.com/CcchenzJ/BootstrapRepresentation.

Specificity-aware Federated Learning with Dynamic Feature Fusion Network for Imbalanced Medical Image Classification.

Recently, federated learning has become a powerful technique for medical image classification due to its ability to utilize datasets from multiple clinical clients while satisfying privacy constraints. However, there are still some obstacles in federated learning. Firstly, most existing methods directly average the model parameters collected by medical clients on the server, ignoring the specificities of the local models. Secondly, class imbalance is a common issue in medical datasets. In this paper, to handle these two challenges, we propose a novel specificity-aware federated learning framework that benefits from an Adaptive Aggregation Mechanism (AdapAM) and a Dynamic Feature Fusion Strategy (DFFS). Considering the specificity of each local model, we set the AdapAM on the server. The AdapAM utilizes reinforcement learning to adaptively weight and aggregate the parameters of local models based on their data distribution and performance feedback for obtaining the global model parameters. For the class imbalance in local datasets, we propose the DFFS to dynamically fuse the features of majority classes based on the imbalance ratio in the min-batch and collaborate the rest of features. We conduct extensive experiments on a dermoscopic dataset and a fundus image dataset. Experimental results show that our method can achieve state-of-the-art results in these two real-world medical applications.

Radioactivity level evaluation in granite building raw material production area in Cenxi, China.

This study was conducted to evaluate the radioactivity level in the granite building raw material production area in Cenxi, China. Natural radionuclide concentrations, γ absorbed dose rates (ADRs) and radon exhalation rates were measured in the area. The spatial distribution of natural radionuclides, γ ADR and radon exhalation rate were mapped with GPS information. The radioactivity levels in the study area were analysed based on the descriptive statistics and frequency distribution of measurement data. According to the Chinese standard, the granite raw materials used for building and decoration in this region were classified based on their radiological hazards. In addition, radiation protection measures were proposed for the mining of local granite building materials products as well as environmental protection around the work area and resident safety.

Federated Domain Adaptation via Transformer for Multi-Site Alzheimer's Disease Diagnosis.

In multi-site studies of Alzheimer's disease (AD), the difference of data in multi-site datasets leads to the degraded performance of models in the target sites. The traditional domain adaptation method requires sharing data from both source and target domains, which will lead to data privacy issue. To solve it, federated learning is adopted as it can allow models to be trained with multi-site data in a privacy-protected manner. In this paper, we propose a multi-site federated domain adaptation framework via Transformer (FedDAvT), which not only protects data privacy, but also eliminates data heterogeneity. The Transformer network is used as the backbone network to extract the correlation between the multi-template region of interest features, which can capture the brain abundant information. The self-attention maps in the source and target domains are aligned by applying mean squared error for subdomain adaptation. Finally, we evaluate our method on the multi-site databases based on three AD datasets. The experimental results show that the proposed FedDAvT is quite effective, achieving accuracy rates of 88.75%, 69.51%, and 69.88% on the AD vs. NC, MCI vs. NC, and AD vs. MCI two-way classification tasks, respectively.

MHW-GAN: Multidiscriminator Hierarchical Wavelet Generative Adversarial Network for Multimodal Image Fusion.

Image fusion technology aims to obtain a comprehensive image containing a specific target or detailed information by fusing data of different modalities. However, many deep learning-based algorithms consider edge texture information through loss functions instead of specifically constructing network modules. The influence of the middle layer features is ignored, which leads to the loss of detailed information between layers. In this article, we propose a multidiscriminator hierarchical wavelet generative adversarial network (MHW-GAN) for multimodal image fusion. First, we construct a hierarchical wavelet fusion (HWF) module as the generator of MHW-GAN to fuse feature information at different levels and scales, which avoids information loss in the middle layers of different modalities. Second, we design an edge perception module (EPM) to integrate edge information from different modalities to avoid the loss of edge information. Third, we leverage the adversarial learning relationship between the generator and three discriminators for constraining the generation of fusion images. The generator aims to generate a fusion image to fool the three discriminators, while the three discriminators aim to distinguish the fusion image and edge fusion image from two source images and the joint edge image, respectively. The final fusion image contains both intensity information and structure information via adversarial learning. Experiments on public and self-collected four types of multimodal image datasets show that the proposed algorithm is superior to the previous algorithms in terms of both subjective and objective evaluation.